MAP: A Model-agnostic Pretraining Framework for Click-through Rate Prediction

TL;DR

This paper introduces a model-agnostic pretraining framework for CTR prediction that leverages self-supervised learning through feature corruption and recovery, significantly improving performance on large-scale datasets.

Contribution

It proposes a novel self-supervised pretraining framework with two algorithms, MFP and RFD, tailored for multi-field categorical data in CTR prediction tasks.

Findings

Achieves state-of-the-art results on Avazu and Criteo datasets.

Enhances model effectiveness and efficiency with the proposed pretraining methods.

Demonstrates compatibility with various backbone models like DCNv2 and DeepFM.

Abstract

With the widespread application of personalized online services, click-through rate (CTR) prediction has received more and more attention and research. The most prominent features of CTR prediction are its multi-field categorical data format, and vast and daily-growing data volume. The large capacity of neural models helps digest such massive amounts of data under the supervised learning paradigm, yet they fail to utilize the substantial data to its full potential, since the 1-bit click signal is not sufficient to guide the model to learn capable representations of features and instances. The self-supervised learning paradigm provides a more promising pretrain-finetune solution to better exploit the large amount of user click logs, and learn more generalized and effective representations. However, self-supervised learning for CTR prediction is still an open question, since current works on this line are only preliminary and rudimentary. To this end, we propose a Model-agnostic pretraining (MAP) framework that applies feature corruption and recovery on multi-field categorical data, and more specifically, we derive two practical algorithms: masked feature prediction (MFP) and replaced feature detection (RFD). MFP digs into feature interactions within each instance through masking and predicting a small portion of input features, and introduces noise contrastive estimation (NCE) to handle large feature spaces. RFD further turns MFP into a binary classification mode through replacing and detecting changes in input features, making it even simpler and more effective for CTR pretraining. Our extensive experiments on two real-world large-scale datasets (i.e., Avazu, Criteo) demonstrate the advantages of these two methods on several strong backbones (e.g., DCNv2, DeepFM), and achieve new state-of-the-art performance in terms of both effectiveness and efficiency for CTR prediction.